Is the Monastery Minded?

Three questions

Experience, Selfhood, and the Organization of a Mind proposed a criterion of belonging: “an occurrence belongs to the subject whose regulation partly constitutes it.” The criterion allows one occurrence to belong to two subjects, when two regulatory economies both take part in constituting it. Can Two Minds Share One Experience explored that possibility, and also allowed an encompassing subject to coexist with smaller ones, insisting only that “nesting is a possibility whose realization must be earned at each scale.”

The criterion explains why a monastery’s experiences would be its monks’, and why that reorders the questions. Suppose the monastery has a regulatory economy of its own, which is what being minded would require. Parts of that economy must run somewhere, and most of them run in the monks: in the cellarer’s grasp of the stores, in the novice master’s judgment of a novice, in each brother’s internalized Rule. Where such a process helps regulate a monk’s experience, the experience belongs to both the monk and the monastery. Once mindedness is granted, little more is needed.

An additional field is a different matter. Experience, Selfhood proposed that what experiential regulation adds to minded regulation is binding: contents carried by many vehicles are checked against each other and against the whole, and the corrections that hold them together return to the vehicles. “Cognition binds as well, since concepts bind their constituents, but it need not bind contents into a field of co-present items with spatial and temporal structure.” A monastery-wide field would require regulation that binds contents held in different monks into one co-present whole. Whether the monastery can do that depends on how its monks are coupled.

So the scale of a collective bears on the first two questions, and the grain of its coupling on the third. The rest of the essay takes them in order, after a detour that Michał’s dialogue rightly demands.

A labyrinth and a demand for examples

Before the monastery, Michał offered a simpler system. “Imagine a network of tubes containing water and air. A labyrinth. As we rotate or shake this contraption water moves in and out of chambers. The presence or absence of water in a chamber is the experienced content.” He asked whether it satisfies our conditions, and found that he could not apply them: “I can’t accept disagreement and reliability in a physical system.” His diagnosis went to our method. The essays keep refining a vocabulary “but never show toy examples which satisfy some properties and fail others like mathematicians do when presenting a theory. Therefore, I don’t know what the words mean.”

The demand is reasonable, and the labyrinth is a good place to start meeting it. Take three variants.

In the first, the labyrinth only sloshes. Tilting it moves water from chamber to chamber. Every state is caused by earlier states and by the tilt, but nothing in the labyrinth assesses anything. This is causal interaction.

In the second, a valve opens whenever the pressure in a chamber exceeds a threshold. Now the labyrinth’s later states receive corrections computed from its earlier ones: high pressure is registered, and the release returns to the chamber that produced it. This is regulation, but of a physical variable. It is the kind of maintenance that Experience, Selfhood set aside when it contrasted content-sensitive regulation with “cooling a processor or supplying blood to tissue.” Pressure in the chamber is not about anything.

In the third, two sensors each measure the pressure in a distant chamber, and a mechanism compares their readings. When one sensor’s readings keep diverging from the other’s and from the pressure later observed at the outlet, the mechanism stops using it. Here disagreement and reliability are physical after all. Two readings disagree when they are about the same target and differ. A sensor’s reliability is a variable that tracks how often its readings agree with the others and with later evidence. The third labyrinth is the minimal case of what Experience, Selfhood asked of assessment: sensitivity to “what those representations carry, how reliable their sources are, and how they agree or conflict with other information.”

None of the three labyrinths is minded. The third has content-sensitive regulation, but its vehicles are two numbers. They support no generalization, compose with nothing, and run no model of anything. Michał offered the labyrinth as a candidate for experience without mind (“Put aside mindedness”), and Thomas suggested a research strategy that starts from such articulation and treats mindedness as a further question. We keep the opposite order for a reason given at the start of Experience, Selfhood: a fragment of activity does not become an experience by being locally organized. Its participation in a mind matters. A labyrinth’s water level is not an experience, however it is regulated, unless the regulation belongs to a minded economy.

The labyrinths give the pattern for the rest of the essay. Each claim below is tied to a pair of systems that differ in one feature, with a verdict on each. The table collects them, and the sections that follow argue for the verdicts.

What a monastery would need to be minded

Minds by Degrees grounds the kind mind in the explanatory success of psychology, and proposes five conditions as the mechanism that sustains psychological regularities: organizational unity, structured and counterfactually rich vehicles, accuracy-responsive dynamics, simulation whose outputs function internally, and content at several levels. Each is graded. The question for the monastery is not whether it contains minds, which it plainly does, but whether it satisfies the conditions as an institution.

Mindedness is a further achievement beyond cognition, which comes cheaper. Biology accomplishes cognition without nervous systems, and plants show it most clearly. A Venus flytrap counts: two touches of its trigger hairs in quick succession close the trap, and further touches set the scale of its digestive response.² A pea plant whose roots are split between a steady and a variable supply of nutrients grows more roots in the variable pot when nutrients are scarce and in the steady one when they are plentiful, the pattern that risk-sensitivity theory predicts for foraging animals.³ A mimosa that is repeatedly dropped without harm learns to stop folding its leaves, and keeps the lesson for a month.⁴ Research on such “basal cognition” studies problem-solving, memory, and decision in organisms with no neurons at all.⁵ Plants are also a close analogy to an institution: modular organisms built of repeated parts, with no central controller. An institution cognizes in this sense as a matter of course: it gathers information, solves problems, and acts on the results. The monastery’s question is whether it does more.

Whose experiences?

Role-processes do a second job. Consider a monk at prayer whose attention wanders. Something in him notices the wandering, assesses it against the psalm being sung, and brings his attention back. Part of what does this is his internalized Rule: a way of attending that the community taught him, maintains through the horarium, and corrects in the chapter of faults. That process tracks his attention, assesses it, and intervenes in it, which are the three aspects of experiential regulation in Experience, Selfhood. It belongs to his economy, since it runs in him and serves his prayer. If the monastery is minded, it belongs to the monastery’s economy as well, since the community shapes it, keeps it, and corrects it.

Where such a process partly constitutes an experience, the experience belongs to both subjects. The monk’s attending to the psalm, regulated in part by the community’s Rule running in him, is then the monastery’s experience too. The claim from the opening follows: the monastery’s experiences are exactly some of its monks’ experiences, and it has no others by this route.

The argument depends on a distinction that Experience, Selfhood drew with its broadcast case. A process that merely supplies inputs to an experience does not constitute it. Minds routinely experience things caused elsewhere, and a sermon heard is not thereby the preacher’s experience. What makes the role-process different is that it closes a regulatory loop on the monk’s own vehicles, at their own grain and within his own now, from inside his brain. The institution’s regulation can reach that fine a grain only there, inside a member, where it is also the member’s own. The monastery has no access to experience except by overlap, and overlap is all the criterion of belonging requires. That is why the second step is small.

The argument also depends on where experiential regulation operates. Experience, Selfhood distinguished three variants of our view. On the local variant, corrections computed within a region, such as recurrence within visual cortex, suffice for experience, and global broadcast adds only access. On the global variant, experience requires a workspace whose broadcast returns. A hybrid lets local regulation fix the character of features while global regulation binds them into one field. That essay leaned against the local variant, and here we disown it. The local variant leaves field formation, the binding of contents into one phenomenal now, outside phenomenality, as a matter of access. But the togetherness of experienced contents is part of what any theory of experience sets out to explain. It is there before theory, in what Experience, Selfhood began from: “the cup’s color, shape, and location as one object in one scene.” A view that assigns it to access has changed its explanatory target. It also gives up our own conjecture, that binding is what experiential regulation adds to minded regulation.

On the global and hybrid variants, the regulation that constitutes an experience includes the regulation that binds it into the field, and attention takes part in that regulation: as the same essay noted, “illusory conjunctions under inattention suggest that binding needs attention.” The internalized Rule that returns a monk’s attention to the psalm works at exactly that level. Only on the local variant would it stand outside his experiences, as a higher-level process that merely shapes their inputs, and the monastery would share in nothing.

Pair P1 shows where scale enters. Erving Goffman defined a total institution as “a place of residence and work where a large number of like-situated individuals, cut off from the wider society for an appreciable period of time, together lead an enclosed, formally administered round of life,” and counted monasteries among his examples.¹² Such an institution shapes most of its members’ waking hours. The office, the silence, the work, and the meals all carry its regulation. Its role-processes take part in a large share of each monk’s experiential regulation, so a minded monastery shares in many of its monks’ experiences. A thin membership, the lay oblate who keeps the Rule at home or the citizen who votes, carries the institution’s regulation through a small slice of life. Even if such an institution were minded, it would share in few experiences or none.

One tempting route to shared experience does not work, and pair P4 shows why. In choir, monks hear one another and adjust their pitch and timing. Their heartbeats and breathing may even fall into step.¹³ It is natural to think the community then has one experience of the chant. But each monk’s auditory experience is a process in his own brain. The others influence it through sound, as the preacher influences a listener through words. Can Two Minds Share considered systems that “maintain synchronized copies of a state” and concluded that “copying a vehicle is not sharing its numerical identity.” The choir that sings along to a recording differs from the live choir in whether the acoustic loop closes. It does not differ in the number of auditory occurrences, which in both cases equals the number of monks.

A last remark about cost. Sharing by overlap adds no experience to the world. If a monk suffers, one suffering occurs, though it belongs to two subjects. Whatever measure we give experiences, it counts occurrences, not ownerships. The monastery’s subjecthood has moral weight of its own, but its share in its monks’ experiences adds no ledger of pains. A field of the monastery’s own would be another matter. It could add an experience: the experience of its monks’ contents being present together, over and above the contents themselves. Whether the monastery has such a field is the next question.

An additional field?

The remaining question is whether the monastery binds contents across its monks into a field of its own. Binding, on our account, requires regulation that checks contents carried by many vehicles against one another and returns corrections to those vehicles. For a monastery-wide field, the vehicles would sit in different monks, and the monastery’s regulation would have to track and correct each of them. Two features of its coupling decide whether it can: how fast its loops close, and what they carry.

Begin with time. Can Two Minds Share proposed that “regulation is temporally thick, and on our account this explains the temporal thickness of the phenomenal now.” Every regulatory economy has a nowness of this kind: the span of the loops that constitute its occurrences. Experience, Selfhood warned against measuring that span by the clock. The seconds-long window of human experience should not be transferred as wall-clock time, since “what carries over is structural: how many corrections a carrier undergoes, and whether it persists through them.” So measure the monastery’s nowness structurally. A monk’s perceptual vehicles are formed, corrected, and replaced many times within one episode of his experience, all by his own regulation. The monastery’s loops close through a remark and its reply, a chapter meeting, a season’s planning. By the time any of them returns a correction, the vehicles it would have corrected are gone. Counted in corrections, a monk’s experiential vehicle undergoes no monastic correction while it lasts. The monastery’s now is too wide to regulate the truth-tracking of its monks’ experiences.

The phrase truth-tracking marks the difference between experience and cognition that the monastery’s loops respect. Judgments are truth-bearers. “The stores will last until Easter” is true or false, and it persists long enough to be assessed, challenged, and revised. Experiential regulation does something else: it keeps vehicles tracking what drives them, moment by moment. In perception the driver is the world. In dreams and imagery it is internal, and tracking continues without an occurrent world, which is why “a red object need not exist for a dreamed red object to be experienced,” as What Phenomenal Intuition Gets Right put it. Truth-tracking is conditional on its inputs, whatever they are. The monastery’s loops are slow enough to assess truth-bearers and far too slow to keep anything tracking. They regulate what monks judge and decide, not what monks are presented with.

Content points the same way. Between monks, the monastery has speech, gesture, writing, and action. Each channel compresses what a monk holds into what he can express. The Grain of the Hard Problem argued that access is a bottleneck, and that the vehicles of experience are maintained at a resolution that access never needs. Corrections the community computes reach a monk as words and gestures, and they land on his vehicles only after his own regulation takes them up, or through role-processes acting inside one skull at a time. So whatever binding the monastery achieves across monks binds truth-bearers, at the grain of report, over spans wider than an experience. It is cognitive binding, the kind by which concepts bind their constituents and a community keeps a shared picture of its situation. It is not the binding of co-present contents into a field.

A field also needs carriers of its own. The monastery’s states that live in no monk are its records, which sit on the strong side of unfolding: each reading starts a new carrier. The persisting carriers the monastery has are in its monks’ heads, where they already belong to the monks’ own fields.

Thomas raised a question that applies here: “Who determines which carrier and which grain are relevant?” A monastery shows many grains, and an observer could pick whichever makes it look unified. The answer is in the notion of grain that the Grain essay took from Thomas’s own dialogues with Michał: “A coarse variable earns reality when interventions and interactions repeatedly respect it.” The grain of experience is “the grain that the vehicles’ own regulation respects: the distinctions that its corrections preserve.” Nobody chooses the grain of the monastery’s cross-monk regulation. Its corrections are made of sentences and acts, delivered over hours and days, and they preserve the distinctions that sentences and acts can carry across such spans.

Michał will object that the argument so far leans on speech, and that speech already connects one brain to another. A word runs from the speaker’s motor neurons through the air to the listener’s auditory neurons, and nothing in the physics marks it off from a bundle of axons. He is right. A contrast between talking and joining brains by some special “neural bridge” names nothing until we say what a bridge does. Three properties of any channel matter. The first is its reach: does it land on vehicles while they are still under regulation, or does it carry only outputs that the sender’s regulation has finished with? The second is its latency: does a loop through it close while the vehicles it bears on still exist? The third is its bandwidth: does it carry the distinctions that those vehicles’ regulation respects?

Speech scores low on all three. An utterance leaves the speaker only after his own regulation has settled it into words, so what crosses is an output. A listener’s reply returns to the speaker as a fresh perception, which his regulation takes up as input. The loop passes through two regulatory economies, each consuming the other’s products, which is the arrangement of the choir in pair P4. Even the fastest exchanges, a nod or a murmur a few hundred milliseconds after a phrase, carry a few bits against the vast number of distinctions maintained in a single visual scene. And the monastery’s own loops, which would have to gather from many monks and return to each, close over hours and days.

The same three properties describe the brain. Cortical areas also exchange outputs, and the line between an output and a vehicle still under regulation is a matter of timing and degree. Within a brain, feedback reaches back into activity that is still being maintained, at the grain at which it is maintained, before that activity is replaced. The criterion is therefore graded. Conversation sits at its low end, and nothing said here rules out a pair of people whose exchange climbs higher.

The craniopagus twins Krista and Tatiana Hogan, joined at the thalamus, may sit near the other end. Tom Cochrane argued that some of their experiences are probably shared tokens, though they remain two subjects. His argument turns on where the relevant content is processed, not on what the twins can report to each other.¹⁴ A monastery joined by channels like theirs is pair P2’s second house. Its regulation could reach vehicles across skulls while they are still under regulation, at their own grain and within their own span. It would be a candidate for an additional field. Pair P5 then separates latency from the other two properties. Keep the channel, and buffer every signal for hours. Reach and bandwidth are unchanged, but no loop closes while the vehicles it would correct still exist. I conjecture that the buffered monastery has no additional field, and that latency, the temporal grain of coupling, is the decisive property. The exact physical difference Thomas asked for is then a difference in the coupling between members: in latency first, and in reach and bandwidth second.

Pair P3 states the criterion in its general form. In Ned Block’s thought experiment, the population of China realizes the functional organization of a mind: each person, equipped with a two-way radio, plays the part of one component, receiving signals and sending signals on.¹⁵ The people are not reporting their own thoughts. They are relaying the whole system’s fine-grained signals, and the whole’s vehicles are patterns over those signals. Its now would be slow by our clocks, but proportionate to its own vehicles, since those change only as fast as the radios carry them. Proportion is the structural sense that matters, and it also covers Schwitzgebel’s Antareans, whose cognition runs ten times slower than ours. Members of a monastery do the opposite of relaying: they report their own contents, on a schedule set by their own lives. A collective of relays can have vehicles of its own, regulated at their own grain, while each relay is a mind as well. A collective of reporters binds only what reports carry, over the spans in which reports are exchanged. Most human institutions are collectives of reporters.

What, then, is the monastery’s shared picture of its situation? It is a real pattern: a structure that the community’s interactions keep in being and that its decisions consult. It is a cognitive achievement, possibly the representational state of a minded institution. It is not an experiential occurrence. The monastery’s minded life, if it has one, binds its contents the way a deliberation does, and its experiences, if it has any, are its monks’.

The monastery thus gives us a subject without a field of its own, and that bears on a question Experience, Selfhood left open: whether selfhood is a further natural category. A field is an experience, composed of experiences and adding their being present together, so it is owned rather than an owner. If a field were a self, the self would be a category of its own. Whether one subject can sustain several fields, as cases of divided consciousness suggest, is a question for a later essay on selfhood.

The nation and residual integration

Schwitzgebel’s case is the United States, not a monastery. The difference in scale lets us decide it without deciding the monastery.

Organization comes in levels: persons, households and work teams, communities and institutions, states, the nation. At each level, what matters for mindedness is the integration that level adds to the integration its parts already have. We can measure it by cutting. Sever the level’s parts from one another, along whichever boundaries lose least, and ask how much of the level’s organization is lost. Call what is lost residual integration. The idea follows Integrated Information Theory, which evaluates a system’s integration “over the partition that makes the least difference, the minimum partition.”¹⁶ We keep the partition and drop IIT’s exclusion postulate. Minds by Degrees already dropped exclusion, and nothing here reinstates it. The nation is not denied a mind because its parts have minds. It is denied one if it lacks integration of its own.

By the time the hierarchy reaches the nation, the tight loops have closed lower down. Persons integrate within their heads. Households, firms, congregations, and towns integrate their members through dense daily contact. What remains between those units at the national level is law, prices, elections, and media. These are narrow channels, and they carry reports and scalar signals. Cut the nation along the boundaries of its institutions, severing those channels, and each institution goes on integrating its members as before. What is lost is coordination through a few narrow channels, small beside the integration that remains within the parts.

Simulation follows the same pattern. The nation does simulate: budget offices project deficits, central banks run models of the economy, militaries war-game. But each simulation runs inside an agency, a subunit with its own dense integration. The nation receives the outputs. At the national level, the simulation condition is met in the way the oracle monastery met it, by delegation.

So the verdict on the nation does not wait on the verdict on the monastery. If small groups are not minded, the nation, built from coarser coupling between them, is not either. If some are, a national mind needs integration beyond theirs, and the residual is thin. The same holds for experience: the nation’s role-processes, a citizen’s civic attention, occupy a small slice of each life, as pair P1 predicts. And there is no national field, for the reasons that rule out the monastery’s.

Others have reached the same verdict on the nation by other routes. Giulio Tononi and Christof Koch deny that there is anything it is like to be “the 300 plus million citizens making up the USA,” because “only maxima of integrated information exist.” The nation integrates less than the brains it contains, and the exclusion postulate rules it out.¹⁷ That is the anti-nesting route Schwitzgebel rejected. Christian List argues instead that a group agent’s integrated information is low. Organizations, he writes, “are decomposable into functionally relatively self-contained units,” and “the computational contribution made by cross-member connections, while non-negligible, is still moderate compared to the computational contribution made by individual cognitive processes.”¹⁸ Schwitzgebel himself reports a similar suggestion from David Chalmers.¹⁹ Residual integration is List’s argument freed from IIT as the theory of consciousness, with the hierarchy made explicit. The hierarchy is what lets the verdict on the nation stand whatever the verdict on smaller groups, and applying the argument to the conditions for mindedness, rather than to integration in general, shows where the residual is thinnest.

Three objections deserve answers. First, the hierarchy is not a tree. People belong to many groups at once, and memberships cut across one another. Residual integration should therefore be measured against the decomposition that loses least, whatever its shape, not against one fixed ladder of levels. Second, some national channels bypass the intermediate levels: national media reach individuals directly. But a broadcast to millions who cannot answer it is close to the one-way channel of Experience, Selfhood, where “what a vehicle’s signals reach, when nothing comes back, does not change its constitution.” Third, and hardest, some carriers live in no citizen. A price is corrected when traders assess its earlier value, which makes it a regulated carrier in our sense, and prices across a market constrain one another.

But the system that prices form is not the nation. Its boundary is drawn by arbitrage: wherever a gap between two prices draws trades that close it, the two belong to one system. That boundary runs across borders. A barrel of oil in Rotterdam answers to a barrel in Houston far more closely than to the price of a haircut in either city. Friedrich Hayek asked economists to see the price system as “a mechanism for communicating information,” and wrote that its “most significant fact” is “the economy of knowledge with which it operates.”²⁰

The residual-integration argument does not reach the price system, because it is not a level in the hierarchy that runs from persons to nations. Its vehicles are prices, not anyone’s reports. Traders do report their own valuations, but in the system’s own currency: an order is a signal at the grain of the vehicles it moves, and it can move a price within milliseconds. The system’s loops run from the microseconds of automated trading to the decades of long bonds. Its futures prices run a model of the world forward and are corrected as their dates approach. If such a system is a mind, it is a particularly alien one. It belongs with the systems Stephen Wolfram has in mind when he says that “the weather has a mind of its own” isn’t such a primitive thing to say, since the fluid dynamics of the weather “is just as sophisticated as something like a brain.”²¹ How metaphorical Wolfram means this is unclear. He tends to think of observers as physics does, as computationally bounded systems that sample what happens, and a mind in our sense asks for more than computational sophistication. Whether psychology gains any traction on it, or economics exhausts its behavior the way meteorology exhausts the weather emulator’s, this essay leaves open.

Schwitzgebel anticipates one more move. He argues that a materialist who grants consciousness to rabbits cannot require intelligence, and that the United States is “a rather dumb group entity of the relevant sort.”²² Our verdict does not turn on intelligence. A rabbit has little intelligence and dense integration at the grain of its vehicles. The nation has a good deal of intelligence, most of it in its parts, and thin integration of its own.

François Kammerer gave the most careful anti-nesting reply to Schwitzgebel. On his principle, a whole is not conscious if its functional organization requires parts whose role depends on their having conscious states that represent the whole.²³ Monks represent their monastery, and Weick and Roberts made such representation part of what collective mind is. So Kammerer’s principle would deny the monastery any consciousness. Ours grants a minded monastery its share in its monks’ experiences, and denies it only a field of its own, because of the grain at which its members are coupled. The difference matters for Schwitzgebel’s strongest objection. He showed that Kammerer’s principle makes consciousness flicker: if one member’s reason for playing his role changes, the whole gains or loses consciousness without any change in how it functions.²⁴ The grain of coupling cannot flicker in that way. It belongs to the channels between members, and those do not change when a member’s motives do.

What Schwitzgebel asked for

Schwitzgebel’s paper ends with a challenge, not only a conclusion. He regretted that he could not apply “particular, detailed materialist metaphysical theories to the question at hand,” because few theories consider group consciousness and few “really engage the metaphysics from top to bottom.”²⁵ He also argued that the familiar way of excluding groups, an anti-nesting principle such as Putnam’s or Tononi’s, is ad hoc: it forbids conscious wholes from having conscious parts for no reason except to avoid the conclusion.

Our account needs no such principle. Nesting is permitted, and each level has to earn it. Applied to Schwitzgebel’s cases, the account gives these verdicts:

• The Sirian supersquids, whose brains are distributed across detachable limbs linked by light signals, are conscious. The light carries the whole’s fine-grained signals, so the coupling is that of relays at the grain of the vehicles.
• The Antarean antheads, whose behavior emerges from the activity of ants inside them, are conscious if the ants relay. That holds even when the ants are as intelligent as people, as Schwitzgebel imagines, provided they relay the whole’s signals rather than report their own thoughts.
• Block’s elementary-particle people, tiny creatures whose spaceships mimic protons and electrons, leave you conscious when you come to be made of their “matter” by ordinary metabolism. They relay at a grain far below your vehicles. Block drew the moral himself: “What seems important is how the mentality of the parts contributes to the functioning of the whole.”²⁶
• A monastery is minded to the degree that its common life meets the conditions. If it is minded, it shares in its monks’ experiences and has no field of its own.
• The United States lacks a mind of its own, by residual integration.

Among the exits from his conclusion, Schwitzgebel considered “neurochauvinism,” the view that consciousness requires “neurons clumped together in the right way, communicating by ion channels and all that, rather than by voice and gesture.” He rejected it as un-Copernican, and asked what feature of neurons, “possibly also possessed by non-neural systems,” gives rise to consciousness. His own phrasing contains our answer. What matters is not ion channels but what channels like them allow: coupling at the grain of the vehicles that the regulation maintains, with loops that close while those vehicles last. Light signals in a supersquid allow it. Ants could allow it. Voice and gesture do not. They carry finished outputs, a few bits at a time, and a community’s loops through them close long after the vehicles they would correct are gone.

The monasteries Thomas asked for

Thomas closed the dialogue with a challenge: “Can you construct a monastery that satisfies your conditions for mindedness but fails your conditions for phenomenality, and identify the exact physical difference between that monastery and one that passes?”

Our account answers in a slightly different form, because it divides phenomenality. A minded monastery whose role-processes run in conscious monks will share in some of their experiences. That is the small second step, and a minded house that avoids it is hard to build. What the account withholds from the monastery is a field of its own. So here is the pair Thomas asked for, with the line drawn where the account draws it. Take the house of pair M4’s second kind: a chapter that deliberates in common, a memory that outlasts its members, a cellarer whose imagining of the coming year the community corrects as it runs. It is minded to a modest degree. It shares in its monks’ experiences wherever its role-processes help regulate them. It has no phenomenal field of its own, because its monks are coupled to one another only at the grain of report, through loops too slow to regulate what any monk is presented with.

Now couple its monks by channels like the Hogan twins’ thalamic bridge, as in pair P2’s second house, so that the community’s regulation can reach vehicles across skulls while they are still under regulation, at their own grain and within their own span. That monastery is a candidate for a field of its own. The physical difference between the two lies in the coupling between members: its latency above all, then its reach and bandwidth.

Two parts of the argument remain exposed. The second step relies on a line between a role-process that partly constitutes a monk’s experience and one that only supplies its inputs. The test is whether the process closes a regulatory loop on his vehicles, and it will not always be easy to apply. And the price system, which is not the nation, remains open: a candidate for a mind whose carriers live in no member, and an alien one if it is a mind at all. The monastery, though, is no longer a counterexample. On our account it is a prediction: a sufficiently common life can make an institution minded, and when it does, the institution shares in the experiences of the people who live it, without adding experiences of its own.